Selective arterial embolization of renal angiomyolipomas: A 10‐year experience

Abstract Objectives To study safety and efficacy of selective endovascular trans‐arterial embolization (TAE) of renal angiomyolipoma (AML) in a 10‐year period at a regional tertiary referral center in Denmark. Patients and methods All 56 patients who underwent TAE of renal AML at Departments of Urology and Radiology, Copenhagen University Hospital – Rigshospitalet, Denmark, from 2009 to 2020 were included. Seven without preoperative and postoperative imaging were excluded, leaving 49 patients for analysis. From national electronic medical records, we retrieved patient characteristics, surgical data, and follow‐up data. Tumor size at the time of embolization and during follow‐up was compared using Student's paired t test. Estimated glomerular filtration rate (eGFR) pre‐ and post‐embolization were compared using Wilcoxon rank sum test. Results We included 49 patients of whom 4 had two tumors treated in the same TAE procedure. Median age was 50 years (interquartile range [IQR]: [29–67 years]), and the median follow‐up time was 4.6 years [IQR: 3.0–6.7 years]. Post‐embolization syndrome (PES) was experienced in 27 patients, and non‐PES in 5 patients. Median length of hospital stay was 0 days [IQR, 0–1]. Postoperative Everolimus immunosuppressive treatment was offered to seven patients. Median tumor size was 6.0 cm [IQR: 4.6–7.9 cm] and was significantly reduced to 3.7 cm [IQR: 2.5–5.2 cm] after treatment (p < 0.001). Kidney function was not affected by TAE. Three deaths, not related to AML, were noted during follow‐up. Conclusion Embolization of AML was in this cohort effective to significantly reduce tumor size without serious adverse events and loss of renal function. TAE is a safe and efficacious treatment and the preferred minimally invasive treatment option of AML.


| INTRODUCTION
Angiomyolipoma (AML) is an uncommon benign renal mesenchymal tumor with female dominance and a prevalence of up to 0.8%. Histologically, it is a mixture of blood vessels, smooth muscle-like cells, and adipose tissue. 1 AML can be diagnosed with ultrasound, computed tomography (CT), or magnetic resonance imaging (MRI), but ultrasound carries a risk of poor specificity for excluding renal cell carcinoma (RCC) as a differential diagnosis. 2 Approximately 80% of AMLs occur sporadically, the remaining due to tuberous sclerosis complex (TSC) or lymphangiomyelomatosis (LAM), both rare genetic conditions. [3][4][5] Due to tortuous aneurysmatic arteries, AML is prone to rupture spontaneously and cause pain, hematuria, retroperitoneal hemorrhage, or even death. Although benign in nature, AML is a concern in pregnancy as well as if lesions are >4 cm where profylactic treatment is preferred due to the risk of rupture. 6 In patients with TSC, AML can be treated with a mammalian target of rapamycin (mTOR) inhibitor (Everolimus). 7 Traditional surgical options include open or laparoscopic removal of the tumor using nephron-sparing surgery or total nephrectomy, the latter in large, complicated tumors. Minimal invasive treatments include thermal ablation or selective endovascular trans-arterial embolization (TAE). Prophylactic surgery reduces the risk of rupture but must carry limited side effects due to the benign nature of the disease. The most frequent treatment of AML is abdominal surgery with its inherent risk of postoperative complications. 8 Selective TAE of renal AML is a minimally invasive treatment with few complications and a tempting option because it is a more nephron-sparing option than surgery. 9 Studies are still limited with short follow-up. [10][11][12][13][14] This retrospective single-center study presents the safety and efficacy on planned and acute selective TAE treatment of AML over a 10-year period at our tertiary referral center. Body mass index (BMI) was grouped into three categories: <25, 25-30, and >30, respectively. Physical status was classified using the American Society of Anesthesiologists (ASA) system. 15 Symptoms at presentation were recorded from the embolization procedures of the 49 patients. Post-embolization syndrome (PES) was defined as fever (above 38.0 C), nausea, and abdominal pain. 16 Other complications were defined as non-PES complications.

| PATIENTS AND METHODS
Tumor size before and after embolization was assessed by an experienced radiologist. The final tumor size after embolization was recorded from the latest available imaging until October 2020. Not all tumors had three-dimensional recordings and thus the maximal diameter on axial CT/MR scans was used to record tumor size. Tumors were F I G U R E 1 (A) Pre-embolization image of a kidney (renal artery and aneurism filled with contrast). (B) Post-embolization image of the same kidney filled with nonspherical polyvinyl alcohol (nsPVA) 355-500 μm and microspheres 250 μm (Embozene ® , Boston Scientific) measured various times throughout the follow-up time. The national laboratory database was reviewed for estimated glomerular filtration rate (eGFR). Preoperative blood tests were recorded as the latest tests available before TAE. Postoperative blood tests were recorded as the latest logged before discharge. If the blood sample was not reported within 2 weeks following TAE, it was regarded as missing.
TAE was performed via the common femoral artery in local or general anesthesia. Selective renal arteriography was performed using a 4 or 5 Fr. diagnostic catheter followed by super-selective catheterization and embolization of the feeding artery (ies) using a microcatheter. Embolization was performed using particulate agents: nonspherical polyvinyl alcohol (nsPVA) 355-500 μm (Bearing nsPVA ® , Merit Medical) or microspheres 250 μm (Embozene ® , Boston Scientific), Figure 1. In selected cases, one or more microcoils (Tornado ® or Nester ® , both Cook Medical) were placed. Most often, this was used in combination with particles in the treatment of microaneurysms.
Puncture site hemostasis was obtained using an Angio-Seal ® (Terumo) and/or manual compression.

| STATISTICAL ANALYSES
Hospitalization days, follow-up time, tumor size, and age were reported as medians with interquartile range (IQR). Nominal variables were reported as counts and percentages. Tumor sizes at the time of embolization and at the end of follow-up were compared using Student's paired t test. eGFR pre-and post-embolization were compared using Wilcoxon rank sum test. A p value of <0.05 was considered statistically significant. Changes in tumor size and changes in the weighted median size in the study period were plotted by linear regression and by the LOESS method. Data analysis was performed in R (version R-3.6.1).

| RESULTS
Of the 49 patients, 39 were females ( Table 1)  Intralesional aneurism >5 mm was identified in connection with five procedures. Angio-Seal ® (Terumo, Europe) vascular closure device was deployed in the femoral common artery for hemostasis, and some patients also needed manual compression.
The median length of stay was 0 days [IQR: 0-1] (Table 3). Postoperative antibiotic treatment was prescribed to eight patients, and postoperative nonsteroid anti-inflammatory drugs (NSAID) treatment in 26 patients, which mirrors the PES diagnosis recorded in 27 procedures. Only one patient was readmitted within 30 days after the TAE procedure, due to PES complications. Non-PES complications were observed in five patients, which included two closure device failures, one kidney abscess, one pseudoaneurysm formation, and one patient treated for pneumonia right after embolization.
Ten patients were re-embolized during the follow-up time due to incomplete response or increasing size of AML. Postoperative Everolimus treatment was offered to seven patients after embolization. The dose was not noted, but compliance problems due to sideeffects were reported.  However, recent studies with robot-assisted laparoscopy suggest that partial nephrectomy of renal masses may be performed as a same-day procedure. 21 In AML, a study by Kara et  There has been a great interest in TAE for the treatment of AML, primarily because it can be performed in local anesthesia and in an outpatient setting. Moreover, tumor size is not a limitation. Our study showed a 40% decrease in tumor diameter, which is comparable with the results found in a systematic review of TAE by Murray et al. 28 The most frequent complication to TAE is PES, ranging from 12.5% in Ramon et al. 29 findings to 56% of cases presented by Fernandez-Pello et al. 8 The incidence of PES was 55% in this study. Long-term follow-up of the EXIST-2 trial shows that Everolimus remains efficacious beyond 4 years of treatment without severe morbidity. 7 In our study, seven patients have prescribed Everolimus after SAE. We did not record the doses taken but did notice that many patients reported compliance problems due to side effects.
This study has strengths and limitations worth mentioning. The strengths include complete and long-term follow-up that captured changes in tumor size and the need for re-embolizations, which has been one of the foremost critiques of TAE. Also, the use of national electronic medical records recorded all readmissions, complications, and long-term mortality, which is important to describe the long-term implication of TAE. The main limitation is the small cohort and retrospective nature of the study that did not include a comparator to other surgical or observational interventions. TAE may be used for patients with a better general health, and this may complicate the comparison with studies of other treatments used for AML. A study of observation versus one or more interventions in the prophylactic treatment of AML is warranted.
In conclusion, our results demonstrate that selective arterial embolization of AML significantly reduced the median tumor size with a low number of severe complications and no significant loss of renal function with a long-term follow-up. TAE is an alternative treatment option for AML and can be considered in patients who are not optimal candidates for abdominal surgery. The most important drawback of TAE is the need for reintervention that occurs in up to 20%. TAE should be compared with surgical treatment in comparative trials, and there is still limited knowledge about factors that predict an optimal response to TAE, which may primarily be related to the material used for embolization.

ACKNOWLEDGMENTS
MAR, SBL, and GN performed research. MAR, SBL, GN, and LL analyzed the data. MAR, SBL, and GN wrote the paper. LL, SH, RJJ, and GN contributed to data gathering for the study. MAR, SBL, and GN designed the research study. All authors revised the paper critically and made contributions writing the paper.